Measures of alcohol-related harm generally focus on health outcomes, particularly mortality. However, as Robin Room points out in his paper in this issue, we tend to forget that society has traditionally been more concerned with acute effects, such as problems of social order, role performance and individual safety, than with the development of chronic alcohol-related disease. This has been very apparent in recent developments in Canada regarding tobacco, where concerns about smuggling, which resulted in few deaths, led to a major cut in taxes, despite evidence of the likelihood of future increases in chronic-disease deaths in the tens of thousands. Another example arises from the discussion preceding the recent provincial ban on smoking on school property. Issues of litter, social control, and traffic safety outweighed health concerns among many legislators and school principals in the province. In the alcohol field, research that suggests reduced risks of moderate drinking compared with life-long abstinence in some populations has provided a renewed impetus to reexamine social harm as an important outcome of alcohol consumption. This issue of Contemporary Drug Problems includes revisions of selected papers on contingencies in the link between drinking and consequences that were originally presented at a conference on Social and Health Effects of Different Drinking Patterns, held in Toronto, Canada, in November 1995. Sponsored by the Addiction Research Foundation, the conference was a thematic meeting of the Kettil Bruun Society for Social and Epidemiological Research on Alcohol. An earlier International Symposium on Moderate Drinking and Health, held in Toronto in April 1993, reviewed the positive and negative health effects of drinking and considered their implications for policy (see Contemporary Drug Problems 21:1 and 21:2, Spring and Summer 1994, for papers from this conference). The earlier symposium had identified as important research issues improvements in the measurement of patterns of drinking, consideration of a wider variety of consequences (social as well as health), and attention to the contingencies of the relationship between drinking and harm in different circumstances and populations.* A major purpose of the 1995 meeting was to address these topics. In this issue, a range of social and health consequences are identified and examined in relation to variations in drinking patterns. David Pittman's paper follows most closely on the 1993 meeting by focusing specifically on the positive effects of moderate drinking on social and physical well-being. He raises the critical issue as to whether moderate consumption-and, we could add, low levels of daily drinking-is a proxy for a generally healthy lifestyle, which includes a variety of other healthful practices. There is good evidence to show that life-long abstainers are atypical, experiencing lower socioeconomic status and less social support. He calls for appropriate differentiations among abstainers and for major improvements in the measurement of alcohol consumption. Paul Gruenewald and his co-authors develop a theoretical framework integrating the role of drinking patterns and drinking behaviors with factors in the risk of alcohol-related injury. …